Women's Thyroid Levels Linked to Risk of Alzheimer's Disease

two generations women, wrapped in blanket

Women with low or high levels of the hormone thyrotropin (also known as Thyroid Stimulating Hormone, or TSH) have an increased risk of Alzheimer's disease. The findings were reported in the July 2008 issue of the medical journal, Archives of Internal Medicine.

In a large study, conducted as part of the community-based Framingham Study, almost 2,000 patients were periodically evaluated for dementia for a period of more than a decade.

Over almost 13 years of follow ups, 209 patients developed Alzheimer's disease. In women, the TSH levels were significantly linked to Alzheimer's disease, and there was a greater than two-fold increased risk of Alzheimer's disease for the women with a TSH below 1.0 and those with levels above 2.1. (Interestingly, no similar relationship was seen in men.)

The researchers do not know whether the thyroid changes are occurring before or after the onset of Alzheimer's disease, or what the connection is, and recommend that further research be conducted to determine if there are any implications for treatment or prevention.

Perhaps the most important finding, however, was that the results were the same, whether patients were taking thyroid medication or not. The operative factor appeared to be the TSH level.

While measurable hyperthyroidism or hypothyroidism can cause cognitive problems, and difficulties with memory, thinking, and learning, it was usually considered to be reversible with proper thyroid treatment.

This study challenges the traditional dogma that thyroid problems are a reversible cause of cognitive impairment, and instead, suggests that an imbalance in thyroid function may even be a contributory factor in Alzheimer's disease.

One of the study's authors, Dr. Zaldy Tan, told Medscape: "It was interesting to find that both low and high levels were associated with Alzheimer's disease.The fact that the brain tries to maintain thyroid levels at a relatively narrow range may suggest that for it to function optimally, it has to be maintained within this range and going below or above that is not a good thing."

It's not clear whether the TSH levels are causing or protecting against Alzheimer's disease, or whether the Alzheimer's is affecting TSH. The study was observational, and did not evaluate causality. It's possible that pituitary responsiveness is damaged by Alzheimer's disease, or that thyroid hormone irregularities may lead to Alzheimer's.

One interesting hypothesis by the study authors looks at the role that thyroid hormone plays in regulating the expression of a gene called the amyloid precursor protein (APP), which has a role in Alzheimer’s. It's possible that imbalances of thyroid hormone may lead to some issues with the regulation of APP, increasing the risk of Alzheimer's. 

Implications for Patients

While further study is needed, this study may end up providing additional reason for the medical community to eventually adopt a narrower TSH reference range. In 2002, the American Association of Clinical Endocrinologists (AACE) recommended narrowing the TSH reference range from the generally used 0.5 to 5.0 mIU/L, to a narrower range of 0.3 to 3.0, with the National Association of Clinical Biochemistry recommending an even lower top limit of 2.5. While some doctors and endocrinologists advocated for the recommendations, the AACE abandoned this recommendation, and laboratories and physicians are still evaluating thyroid blood tests according to the old 0.5 to 5.0 range.

In the meantime, however, this study provides further evidence to argue that, for patients receiving thyroid treatment, the target range for optimal TSH should be between 1.0 to 2.0, unless other more immediate factors are involved. (For example, some thyroid cancer patients are maintained on suppressive doses of thyroid medication, keeping TSH low -- and sometimes near 0 -- to prevent cancer recurrence.)


Tan, Zaldy et. al. "Thyroid Function and the Risk of Alzheimer Disease: The Framingham Study." Archives of Internal Medicine, 2008;168(14):1514-1520.

Continue Reading